International Business Machines Corporation

Qi Liang

Gui Yu Jiang

Tian Tian

Weixiong Rao

Qin Yue Chen

Qi Ye

Guohua Li

Yi Xuan Zhang

Donald W Schmidt

Feng Liu

Daniel V Rosa

Hong Lin

Chao Wei

Xiao Ling Chen

Yu Sun

Hong Min

Heng Wang

Shuai Wang

Gong Su

Yi Xu

Xue Yong Zhang

Long Cheng

Tie Liu

Feifei Li

Yi Fang

Yong Xin Qi

Pu Yp Yang

Chang Xin Miao

Su Han

Liu Xin

Zhe Dong Mo

Peng Zhao

Data format conversion processing of an accelerator accessed by a processor of a computing environment is reduced. The processor and accelerator use different data formats, and the accelerator is configured to perform an input conversion to convert data from a processor data format to an accelerator data format prior to performing an operation using the data, and an output conversion to convert resultant data from accelerator data format back to processor data format after performing the operation. The reducing includes determining that adjoining operations of a process to run on the processor and accelerator are to be performed by the accelerator, where the adjoining operations include a source operation and destination operation. Further, the reducing includes blocking an output data format conversion of the source operation and an input data format conversion of the input data for the destination operation.

Reducing data format conversion of an accelerator

A computer-implemented method, a computer system, and a computer program product for trajectory similarity search is provided. The present invention may include, in response to receiving, by one or more processors, a search request for at least one trajectory similar to a query trajectory, determining, by one or more processors, a respective similarity between a query trajectory and a plurality of trajectories by calculating, in a synchronized way, a spatial distance measure and a time difference measure between the query trajectory and the plurality of trajectories. The present invention may further include, identifying, by one or more processors, the at least one trajectory from the plurality of trajectories based on the respective similarity between the query trajectory and the plurality of trajectories.

Trajectory similarity search

Data format conversion processing of an accelerator accessed by a processor of a computing environment is reduced. The processor and accelerator use different data formats, and the accelerator is configured to perform an input conversion to convert data from a processor data format to an accelerator data format prior to performing an operation using the data, and an output conversion to convert resultant data from accelerator data format back to processor data format after performing the operation. The reducing includes determining that adjoining operations of a process to run on the processor and accelerator are to be performed by the accelerator, where the adjoining operations include a source operation and destination operation. Further, the reducing includes identifying for removal output data format conversion of output data of the source operation for input to the destination operation, and input data format conversion of the input to the destination operation.

A source code defining record structure of data is received. The source code includes data tags which provide information for processing to be performed on the data. Data tagging metadata is generated based on the source code, and the data tagging metadata includes record structure information and the data tags. And data is processed according to the data tagging metadata.

Data processing with tags

A computer-implemented method for proactive change verification is provided. Aspects include analyzing runtime execution characteristics from a plurality of base activity metrics and a plurality of experimental activity metrics and creating a plurality of activity pairs wherein each activity pair comprises a base activity metric and its corresponding experimental activity metric. Aspects also include identifying significant activity pairs from the plurality of activity pairs wherein the experimental activity significantly deviates from its corresponding base activity and classifying significant activity pairs by deviation type. Aspects further include sorting activity pairs by their impact and removing activity pairs that have an impact of less than a threshold amount, wherein the threshold amount is based on the deviation type. Aspects also include correlating interacting impactful activity pairs temporally and prioritizing impactful activity pairs using experimental activity deviations and displaying the impactful activity pairs.

Proactive change verification

Techniques for profile clustering for homogenous instance analysis are described herein. An aspect includes collecting for each of a plurality of homogenous instances in a cloud computing environment, respective profiling information, the profiling information comprising counter data. Another aspect includes identifying a code unit for analysis. Another aspect includes aggregating counters corresponding to the code unit for each of the plurality of homogenous instances from the respective profiling information. Another aspect includes determining a cluster comprising a subset of the plurality of homogenous instances based on the aggregated counters.

Profile clustering for homogenous instance analysis

In one embodiment of the present disclosure, an original graph including nodes is obtained. The nodes of the original graph are reordered to generate a reordered graph. Non-zero elements in an adjacency matrix for the reordered graph are clustered as compared with an adjacency matrix for the original graph. The adjacency matrix for the reordered graph is encoded with integers. The integers correspond to non-empty blocks in the adjacency matrix for the reordered graph.

Directed graph compression

A method, computer system, and computer program product for accelerating class data loading in a containers environment are provided. In response to a first container in a containers environment being created from a first image, at least one archive file containing a set of classes from the first image can be loaded. Then a respective class sharing file for each of the at least one archive file can be generated. The class sharing file is stored in a shared location. A second container in the containers environment is created from a second image. If a class sharing file from the archive is found in the shared location, that class sharing file can be used.

Class data loading acceleration

Providing web search results which are verifiable via the blockchain may offer transparency into valid web pages and their availability when conducting a search. One example method of operation may include performing a plurality of web page interactions by a plurality of network node devices operating on a network, validating existence of valid web pages, detecting invalid web pages based on the plurality of web page interactions, and storing a valid page package with the valid web pages in a blockchain.

Trustable web searching verification in a blockchain

Embodiments of the present invention provide a method, a system, and a computer program product for disassembly code generation during debugging. In the method, a first disassembly code is obtained based on a source code of a computer program. Runtime information of the computer program is obtained from a debugging engine. And a second disassembly code is generated based on the first disassembly code and the runtime information. Computing works about generating disassembly code are distributed in different phases. When the user requests a view of the disassembly code, the response time will be reduced.

Disassembly code generation during debugging

Processing instruction grouping information is provided that includes: reading addresses of machine instructions grouped by a processor at runtime from a buffer to form an address file; analyzing the address file to obtain grouping information of the machine instructions; converting the machine instructions in the address file into readable instructions; and obtaining grouping information of the readable instructions based on the grouping information of the machine instructions and the readable instructions resulted from conversion. Status of grouping and processing performed on instructions by a processor at runtime can be acquired dynamically, such that processing capability of the processor can be better utilized.

Code optimization based on information of readably converted executed instruction groups represented in address file

A method for control flow analysis according to an embodiment of the present invention includes: acquiring an original function call tree of a program, wherein nodes of the original function call tree represent functions and a parent/child relation between the nodes represents a calling relation; generating a corresponding function dominator tree from the calling relation, wherein nodes of the function dominator tree represent the functions and a parent/child relation between the nodes represents a dominator relation, wherein a first function dominates a second function if all the invocations to the second function are originated by the first function; and simplifying the original function call tree according to the function dominator tree so as to obtain a simplified function call tree. According to an embodiment of the present invention, the function call tree for control flow analysis can be simplified.

Control flow analysis utilizing function dominator trees

The image data compression method involves the performing of steps: a block dividing step for dividing a computer processor pipeline statistic image to be displayed into a plurality of blocks with predetermined block width and block height; a block list creating step for creating a block list for indexed blocks containing meaningful pixels in the divided blocks such that the indexed blocks are associated with each other, wherein the meaningful pixels are pixels corresponding to the computer processor pipeline raw trace data; and a pixel information compressing step for compressing original pixel information on the meaningful pixels in the indexed blocks and storing it in the block list. Apparatus for performing the method is also provided.

Growing community of inventors

Shanghai, China

Qi Liang